About Diffuse Bio
At Diffuse Bio we’re building a push-button entirely AI software platform for drug design, leveraging breakthroughs in generative AI. Our team has been behind breakthroughs in AI protein design for the past 6 years, including the first experimental validation of AI-generated proteins and diffusion models for protein structure and sequence.
About the role
Skills: Machine learning, Python, Deep LearningThe role:
- Extend and scale Diffuse's in-house deep generative modeling toolkit for
downstream applications in molecular design. - Thoughtfully execute deep learning experiments to improve performance
of models or develop new functionality (e.g. loop engineering,
structure prediction of protein-protein complexes). - Work closely with software engineers to build systems for efficient
training and deployment of deep learning models.
Ideal background:
- Self-starter who enjoys working on tough scientific problems and is results-driven.
- Able to think critically, methodically, and creatively about experiments.
- Proficient in Python.
- Experience working with deep learning frameworks (e.g., PyTorch).
- 3+ years of industry experience in a data science or engineering position.
- Track record of impressive work in industry/academia centered on ML / deep learning.
- Graduate degree in math, CS, stats, bioengineering, comp bio, or a related field (not a hard requirement for exceptional candidates).
- Is located in the Bay Area (remote work is an option for exceptional candidates).
Pluses:
- Knowledge of physics, math, molecular biology, chemistry, etc.
- Previous work on ML applied to problems in structural biology or molecular design.
- Strong publication record.
What we offer:
- The opportunity to join the founding team and play a critical and expanding role in shaping the company.
- The opportunity to work on cutting-edge AI with leading researchers from top institutions.
Technology
We’re building cutting edge models and software for generative molecular design, and our primary focus is innovating on the algorithms and approaches. We’re also building a seamless software product to deploy our models internally and to external customers/partners and back-end solutions to enable rapid model inference and distributed training.